Pinned Tweet
I’m super excited to share that I’ve joined @physical_int as a Research Scientist!
14
4
231
12,778
After LBM 1 and VLA Foundry, I wanted to find a place to keep pushing on the science of generalist robot learning at larger scales. I’ve really been struck by both the scale and thoughtfulness (these two qualities can be difficult to maintain at the same time!) here at Pi.
1
0
4
394
Image augmentation is widely used to make learned vision algorithms more robust. MobileVISTA is the natural follow up to our previous work VISTA, exploring how to augment demos in the image space for greater robustness to changes in robot base positions.
Mobile manipulation policies can break from base pose errors of just a few cm, which are common after navigation. Can we get pose generalization without additional demos? Introducing MobileVISTA: collect demos at one base pose, get a policy that works from many 🧵👇
Website: sdwistreich.github.io/mobilevista
Paper: arxiv.org/abs/2610.07511
0
1
10
1,595
Katherine Liu retweeted
Proud to debut Walden Robotics, and share we've raised $300M in funding co-led by Toyota and Deviation Capital, with participation from NVIDIA, Boeing Ventures, Samsung, and others.
We're building general-purpose robots that continuously learn while performing real work, alongside people. Today.
youtube.com/watch?v=fewvZr…
We're building general-purpose robots that continuously learn while performing real work, alongside people. Today.
youtube.com/watch?v=fewvZr…
6
19
93
18,351
I’ll be at ICRA this year! Excited to visit Vienna and catch up with everyone. I’ll be giving a talk at the VLA Pipelines Workshop on Friday afternoon. Looking forward to talking robots :)
0
0
3
326
Katherine Liu retweeted
Submit your CoRL workshop proposal! This year @RLioutikov and I wanted to make the workshop more "workshopy".
Main changes are:
- Half-day events only
- Limited speaker slots
- Challenge- and participation-driven
- A post-workshop artifact (white paper, report, paper, etc.) summarizing the discussions
Main changes are:
- Half-day events only
- Limited speaker slots
- Challenge- and participation-driven
- A post-workshop artifact (white paper, report, paper, etc.) summarizing the discussions
📣 CoRL 2026 Call for Workshops & Tutorials is live!
This year: workshops should be workshops again, not mini-conferences. We want formats where the community actually works together.
Half-day events | Nov 9 | Austin, TX 🤠
Details: corl.org/contributions/…
#CORL2026
0
7
37
8,101
Katherine Liu retweeted
Releasing RecGen: a collaboration between @ToyotaResearch, @toyota_europe, and @UvA_Amsterdam tackling a core 3D vision challenge: reconstructing complete multi-object scenes (parts, poses, textures, even occluded geometry) from just 1 to a few RGB-D views.
Trained purely on achieves SOTA on real-world robotics and 6D pose benchmarks, handling occlusions, symmetry, and complex interactions.
A step toward scalable, high-fidelity digital twins for robotics, and better evaluation and training of generalist policies.
reconstruction-by-generation.github.io
Trained purely on achieves SOTA on real-world robotics and 6D pose benchmarks, handling occlusions, symmetry, and complex interactions.
A step toward scalable, high-fidelity digital twins for robotics, and better evaluation and training of generalist policies.
reconstruction-by-generation.github.io
3
35
222
27,635
Katherine Liu retweeted
a good model trained on even a simple task with a tiny amount of data feels mesmerising, no matter how many times you see it.
15
14
163
23,275
Katherine Liu retweeted
Gave a talk on Robot Learning from Human Data at Stanford. It was great to be back!
Some opinionated points:
1. Human data collection capacity is outpacing the research.
2. We still don't have the "science" for scaling robot capability with human data.
3. We are far from being tic human behaviors.
youtube.com/watch?v=NUtaN1…
Some opinionated points:
1. Human data collection capacity is outpacing the research.
2. We still don't have the "science" for scaling robot capability with human data.
3. We are far from being tic human behaviors.
youtube.com/watch?v=NUtaN1…
1
21
180
15,047
Katherine Liu retweeted
[5/5] But wait, there’s more! Bring your own datasets, load from LeRobot, visualize everything, and explore all tutorials 🧩: github.com/TRI-ML/vla_fou…
Project website: tri-ml.github.io/vla_foundry/
Project website: tri-ml.github.io/vla_foundry/
0
1
1
104
Katherine Liu retweeted
[4/5] Grand finale: evaluate your VLA Foundry robot policies in simulation on 40+ challenging manipulation tasks! 🤖: github.com/TRI-ML/vla_fou…
1
1
2
108
Katherine Liu retweeted
[3/5] Load our released LLMs and VLMs from HuggingFace, and do inference on your own data ⚡(text completion & image captioning): github.com/TRI-ML/vla_fou…
1
1
1
89
Katherine Liu retweeted
[2/5] Train the entire LLM → VLM → VLA pipeline, all on one place🧠: github.com/TRI-ML/vla_fou…
1
1
1
133
Katherine Liu retweeted
[1/5] Want to build your own LLM → VLM → VLA models without the usual headache? VLA Foundry makes it *super* easy. 🛠️ Get started with self-contained tutorial notebooks that anyone can run: tri-ml.github.io/vla_foundry/do…. Examples in 🧵
2
1
5
227
A few interesting rollouts from the Foundry-QwenVLA-2.5B multi-task model on seen tasks in sim – a 🧵. I really like behaviors that involve non-prehensile manipulation, like the little nudges in StoreCerealBoxUnderShelf.
Releasing VLA Foundry: an open-source framework that unifies LLM, VLM, and VLA training in a single codebase. End-to-end control from language pretraining to action-expert fine-tuning — no more stitching together incompatible repos.
2
20
117
15,213
Katherine Liu retweeted
In LBM Eval, our Drake-based multi-task tabletop manipulation simulation:
• Our from-scratch VLA matches our prior LBM
• Our VLA using a Qwen3-VL backbone tops that by over +20%
• Our from-scratch VLA matches our prior LBM
• Our VLA using a Qwen3-VL backbone tops that by over +20%
2
1
7
1,855
Katherine Liu retweeted
VLA Foundry supports both from-scratch training and pretrained HF backbones. We trained two model types to show it off:
• Fully from scratch via our LLM→VLM→VLA pipeline
• Built on Qwen3-VL
We share the weightshttps://huggingface.co/collections/TRI-ML/vla-foundry
• Fully from scratch via our LLM→VLM→VLA pipeline
• Built on Qwen3-VL
We share the weightshttps://huggingface.co/collections/TRI-ML/vla-foundry
1
2
13
2,315
Also, if you’re wondering how we generated all these cool videos from the Drake sim, check out @NicholasEPfaff’s repo github.com/nepfaff/drake-… as a starting point 👀
A few interesting rollouts from the Foundry-QwenVLA-2.5B multi-task model on seen tasks in sim – a 🧵. I really like behaviors that involve non-prehensile manipulation, like the little nudges in StoreCerealBoxUnderShelf.
0
6
28
3,006
Katherine Liu retweeted
Having control over upstream LLM/VLM training is key to training a good robotics model.
We hope VLA Foundry opens the door for researchers and practitioners to answer questions they previously wouldn’t even have thought of asking if upstream pretraining was simply inherited!
We hope VLA Foundry opens the door for researchers and practitioners to answer questions they previously wouldn’t even have thought of asking if upstream pretraining was simply inherited!
Releasing VLA Foundry: an open-source framework that unifies LLM, VLM, and VLA training in a single codebase. End-to-end control from language pretraining to action-expert fine-tuning — no more stitching together incompatible repos.
2
6
28
3,588